AI and cognitive psychology rant (getting more and more OT - tell me if I should shut up)

[John J. Lee]

Stephen Horne <steve at ninereeds.fsnet.co.uk> writes:

> On 26 Oct 2003 20:56:09 +0000, jjl at pobox.com (John J. Lee) wrote:
> 
> >> Yes, but why can we see the affects of superposition at the
> >> microscopic scale but not at the macroscopic. That is what strikes me
> >> as odd - if parallel universes work as an explanation, then why do
> >> they work differently at the two scales. In particular, why can we not
> >
> >They don't.
> 
> Are you claiming that in schroedingers experiment that the dead and
> live cats interact in some way that can be measured outside the box
> without collapsing the waveform?

Well, in the many-worlds interpretation (MWI) there *is* no
wavefunction collapse: everything just evolves deterministically
according to the Schrodinger equation.  But of course, since cats are
big lumps of matter, one wouldn't expect to be able to measure
interference effects using cats.


> I was also under the impression that the largest 'particle' to be
> successfully superposed in an experiment was a buckyball (or something
> like that - at least a 'large' molecule of some kind or another) and
> the timescale for that superposition was tiny.

The largest *measured* superposition, yes.  The Copenhagen
interpretation says that the world evolves according to the
Schrodinger equation until, um, it stops doing that, and collapses to
an eigenstate.  When does the Copenhagen interpretation say the wfn
collapses?  It doesn't!  It denies any meaning to that question.
That's claiming that we just "shouldn't" ask about this part of
reality, and stop our enquiry there.  Why should I follow that
instruction when the MWI explains exactly what happens?  If a theory
explains more than its rival, one rejects the rival theory.  And it
doesn't make any sense to say "there are many universes, except for
large objects, for which there is only one universe".  This brings us
into epistemological issues which Deutsch deals with in his book much
better than I can.

Of course, there's more to this debate than Copenhagen vs. MWI, but
the other rival theories all (to my very limited knowledge) seem to be
either re-hashings of MWI in disguise, or complicated theories that
introduce ad-hoc irrelevancies without any compensating benefit.  And,
to dispense with the absurd objection that MWI is 'expensive in
universes', since when has complexity of *entities* been a criterion
on which to judge a theory??  Complexity of *theories* of the world is
a problem, complexity of the world itself is not.  Indeed, one thing
we know independent of any theory of quantum mechanics (QM) is that
the world is damned complicated!


> Yet the whole point of the thought experiment is that according to the
> theory, as conventionally described (I know next to nothing of the
> detail), it should be possible for a cat to be superposed almost as
> easily as it is possible for a subatomic particle - a simple
> cause-and-effect chain is all that is needed. If that is the case,
> superpositions of macroscopic objects should be happening all the
> time.

They do, yes!


> Now either the superpositions are in parallel universes with each
> state undetectable from an observer in another one of those universes,

*The fact that those superpositions exist* is justified by the fact
that MWI is the best theory of QM that we have.  The particular nature
of a particular large object's superposition is not measurable.
Contrary to popular belief, this raises no major epistemological
problems for MWI, and does not turn it into metaphysics.


> or they are in the same universe and detectable in some way, or there
> is a differentiation between the microscopic and macroscopic scales,
> or - and this is very likely, I admit - I am seriously confused about
> what the hell is going on (the natural state for a human confronted
> with quantum theory).

Saying that superpositions are "in one universe" or another seems to
be playing mix-n-match with the various theories.

[...]
> >We see exactly the effects that the theory predicts.  They're just
> >very small.
> 
> OK - so why is it not possible to detect the superposition of that
> cat? Why is the experiment still considered a thought experiment only?

Simply because that's what QM predicts for large objects.  The
'accident' of the size of Planck's constant means that interference
effects are small for large objects.  The universes involved are none
the less real for that: denying that requires doublethink.
Interference effects aside, why *should* we experience anything
unusual when "we" (scare quotes because issues of personal identiy
come up here, of course) exist as a superposition, ie. when we exist
in multiple universes?  There is a very close parallel here with
people's disbelief in the round-earth theory because they couldn't see
why they wouldn't fall off the earth if they moved too far from "the
top of the earth".  Why don't we fall off the earth?  Because the
(scientifically justified) theory says we won't.  Why doesn't the me
in this universe experience multiple universes simultaneously?
Because the (scientifically justified) theory says I won't.  Why
*should* we experience multiple universes?  -- universes are entirely
independent of each other apart from interference effects that are
only large for very small objects, or slightly larger and very
carefully constructed ones.

But again, for those arguments in more detail you're vastly better
advised to go to David Deutsch's (extremely readable and enlightening)
book than to me :-)

[...]
> >Yeah -- hence the solipsism joke.
> 
> Ah - sorry - I'm not actually familiar with that term.

Well, explaining a joke always spoils it, but: a solipsist is a person
who believes that he is the only real thing in existence.  The rest of
the universe, to a solipsist (to *the* solipsist, in fact ;-) is
simply the result of his own imaginings.  Deutsch very clearly
presents an argument that this position is indefensible and
meaningless, starting from that joke I quoted.


John